image_crawl/run.py

#!/usr/bin/env python
if __name__ != '__main__': raise Exception("Do not import me!")

from argparse import ArgumentParser
from google_images_download import google_images_download   #importing the library

parser = ArgumentParser()

parser.add_argument("classnames",
	help="file containing a class name in each line")

parser.add_argument("--output_directory", "-o", default="download",
	help="output folder")

parser.add_argument("--limit", "-l", default=60, type=int,
	help="number of images to dowload")


def main(args):

	with open(args.classnames, "r") as f:
		names = f.readlines()

	query_names = [name.strip().partition(".")[-1].replace("_", " ").lower() for name in names if not name.startswith("#")]
	# query_names = [name.strip() for name in names if not name.startswith("#")]

	response = google_images_download.googleimagesdownload()   #class instantiation

	print(f"Found {len(query_names)} query names")
	paths = response.download(dict(
			keywords=",".join(query_names),
			limit=args.limit,
			print_urls=False,
			size=">800*600",
			output_directory=args.output_directory,
			chromedriver="chromedriver"
		))   #passing the arguments to the function

	# print(paths)   #printing absolute paths of the downloaded images





main(parser.parse_args())